Fix assert with undeliverable message from comm actor #1606
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
meta-cla
bot
added
the
CLA Signed
|
@dulinriley has exported this pull request. If you are a Meta employee, you can view the originating Diff in D84952942. |
dulinriley
added a commit
to dulinriley/monarch
that referenced
this pull request
Oct 17, 2025
…orch#1606) Summary: Fix a panic in PythonActor::handle_undeliverable_message when the "sender" is the comm actor. We need to update the sender back to the original "self" actor by using the headers set in the envelope. After this fix, instead of a panic we get a supervision error: ``` SupervisionError( Actor ...wrapper_1xYczTZiTdb1[0] exited because of the following reason: <PyActorSupervisionEvent: ...wrapper_1xYczTZiTdb1[0]: failed: serving ...wrapper_1xYczTZiTdb1[0]: processing error: a message from ...wrapper_1xYczTZiTdb1[0] to ...fail_1am38hE5fnus[0] was undeliverable and returned: Some("send error: channel closed; multicast error: comm actor comm_1JgvjFbdpnUf[0] failed to deliver the cast message to the dest actor; return to its original sender's port ...wrapper_1xYczTZiTdb1[0] ) ``` Not very terse, but better than a panic! This also allows any custom override of handle_undeliverable_message to work. Reviewed By: pablorfb-meta Differential Revision: D84952942
dulinriley
force-pushed
the
export-D84952942
branch
from
6cfcbc1 to
730c845
Compare
dulinriley
changed the title
Contributor
Author
|
Updated to actually fix the problem instead of just enhancing the assert |
) Summary: Fix a panic in PythonActor::handle_undeliverable_message when the "sender" is the comm actor. We need to update the sender back to the original "self" actor by using the headers set in the envelope. After this fix, instead of a panic we get a supervision error: ``` SupervisionError( Actor ...wrapper_1xYczTZiTdb1[0] exited because of the following reason: <PyActorSupervisionEvent: ...wrapper_1xYczTZiTdb1[0]: failed: serving ...wrapper_1xYczTZiTdb1[0]: processing error: a message from ...wrapper_1xYczTZiTdb1[0] to ...fail_1am38hE5fnus[0] was undeliverable and returned: Some("send error: channel closed; multicast error: comm actor comm_1JgvjFbdpnUf[0] failed to deliver the cast message to the dest actor; return to its original sender's port ...wrapper_1xYczTZiTdb1[0] ) ``` Not very terse, but better than a panic! This also allows any custom override of handle_undeliverable_message to work. Reviewed By: pablorfb-meta Differential Revision: D84952942
dulinriley
force-pushed
the
export-D84952942
branch
from
730c845 to
406195c
Compare
|
This pull request has been merged in 1a3442a. |
AlirezaShamsoshoara
pushed a commit
to AlirezaShamsoshoara/monarch
that referenced
this pull request
Oct 30, 2025
) Summary: Pull Request resolved: meta-pytorch#1606 Fix a panic in PythonActor::handle_undeliverable_message when the "sender" is the comm actor. We need to update the sender back to the original "self" actor by using the headers set in the envelope. After this fix, instead of a panic we get a supervision error: ``` SupervisionError( Actor ...wrapper_1xYczTZiTdb1[0] exited because of the following reason: <PyActorSupervisionEvent: ...wrapper_1xYczTZiTdb1[0]: failed: serving ...wrapper_1xYczTZiTdb1[0]: processing error: a message from ...wrapper_1xYczTZiTdb1[0] to ...fail_1am38hE5fnus[0] was undeliverable and returned: Some("send error: channel closed; multicast error: comm actor comm_1JgvjFbdpnUf[0] failed to deliver the cast message to the dest actor; return to its original sender's port ...wrapper_1xYczTZiTdb1[0] ) ``` Not very terse, but better than a panic! This also allows any custom override of handle_undeliverable_message to work. Reviewed By: pablorfb-meta Differential Revision: D84952942 fbshipit-source-id: b2cb36600ca89a03e7a85cfccd46ce6bcf2487cf
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary:
This panic is very common due to the CommActor message forwarding which obscures
the return address.
When this happens, have a better message that also includes what the destination was
so people aren't confused.
Reviewed By: pablorfb-meta
Differential Revision: D84952942